Bram Wasti dc48ba0c75 Kernel-override Determinism [1/n] (#25603) ...

Signed-off-by: Bram Wasti <bwasti@meta.com>

2025-09-26 16:59:09 -07:00

..

mla

[V0 deprecation] Remove _VLLM_V1 suffixes from attention backend names (#25489)

2025-09-25 17:37:50 +00:00

__init__.py

[V1] Implement vLLM V1 [1/N] (#9289)

2024-10-22 01:24:07 -07:00

cpu_attn.py

[V0 deprecation] Remove _VLLM_V1 suffixes from attention backend names (#25489)

2025-09-25 17:37:50 +00:00

flash_attn.py

[V0 deprecation] Remove _VLLM_V1 suffixes from attention backend names (#25489)

2025-09-25 17:37:50 +00:00

flashinfer.py

[V0 deprecation] Remove _VLLM_V1 suffixes from attention backend names (#25489)

2025-09-25 17:37:50 +00:00

flex_attention.py

Kernel-override Determinism [1/n] (#25603)

2025-09-26 16:59:09 -07:00

gdn_attn.py

[Qwen3-Next][GDN] fixes cuda graph capturing bug in GDN metadata and a stride bug in causal_conv_1d. (#25743)

2025-09-26 01:18:58 -07:00

linear_attn.py

[Spec Decode] Enable FlashInfer Spec Decoding (#25196)

2025-09-23 22:29:58 -04:00

mamba1_attn.py

[Attention] FlashAttn MLA (#14258)

2025-09-04 02:47:59 -07:00

mamba2_attn.py

[Model] Mamba2 varlen refactor (#21467)

2025-09-26 11:31:14 +00:00

mamba_attn.py

[Spec Decode] Enable FlashInfer Spec Decoding (#25196)

2025-09-23 22:29:58 -04:00

pallas.py

[V0 deprecation] Remove _VLLM_V1 suffixes from attention backend names (#25489)

2025-09-25 17:37:50 +00:00

rocm_aiter_fa.py

[V0 deprecation] Remove _VLLM_V1 suffixes from attention backend names (#25489)

2025-09-25 17:37:50 +00:00

rocm_attn.py

[V0 deprecation] Remove _VLLM_V1 suffixes from attention backend names (#25489)

2025-09-25 17:37:50 +00:00

short_conv_attn.py

[Spec Decode] Enable FlashInfer Spec Decoding (#25196)

2025-09-23 22:29:58 -04:00

tree_attn.py

[V0 deprecation] Remove _VLLM_V1 suffixes from attention backend names (#25489)

2025-09-25 17:37:50 +00:00

triton_attn.py

[V0 deprecation] Remove _VLLM_V1 suffixes from attention backend names (#25489)

2025-09-25 17:37:50 +00:00

utils.py

Remove cuda hard-code in compute_causal_conv1d_metadata (#25555)

2025-09-26 01:19:20 -07:00

xformers.py

[V0 deprecation] Remove _VLLM_V1 suffixes from attention backend names (#25489)

2025-09-25 17:37:50 +00:00